Background/Purpose: TNF is a key pro-inflammatory cytokine in normal immune responses and pathologically in the rheumatoid arthritis (RA) synovium, thus representing a key treatment target but one where potential negative effects on responses to natural infection and vaccination must be balanced. We hypothesized that some of the clinical efficacy of TNF blockade in RA may be mediated by effects on the B cell compartment and have previously reported that RA patients on anti-TNF (etanercept) have a disruption of germinal center (GC) reactions in peripheral lymphoid tissue (Anolik, J Immunol., 2008, 180: 688). Here, we characterized vaccine responses in RA patients recently started on TNF blockade.

Methods: We randomized subjects in a 2:1 ratio to receive standard dosing regimens of etanercept or adalimumab for 24 WK. The subjects met the 1987 ACR criteria for RA, clinically active (DAS28>4.4), and were on MTX. Disease activity and response was assessed based on DAS28-CRP.  Twenty eight of 63 randomized subjects participated in a vaccine sub-study receiving vaccination with Hepatitis A (accelerated 12 and 16 WK), Hepatitis B (accelerated 20 mg at 12, 16, 20 WK), diphtheria/tetanus (dT) booster (12 WK) or combination. Response to Hep B was defined as an antibody titer ≥12 up to 36 WK after randomization. Response to Hep A was defined as a detectable antibody at any point after immunization (to 20 WK). dT booster responder was defined as a titer  > 1.0 IU/mL and a ratio to pre-boost of > 3.0. Flow cytometry of B cells was performed at baseline and at 12 and 24 WK. Expression of CD19, IgD, CD27, CD38 and CD24 were used to identify B cell subsets.

Results: A surprisingly high fraction of patients receiving Hep B vaccination were non-responders (NR) (78%, n=18/23). All 5 patients who responded to Hep B were receiving etanercept. Responses to Hep A were higher (48%, n=10/21). Two out of 5 patients who received dT vaccine were responders. For patients who received both Hep A and Hep B vaccinations (n=14), 21% responded to both (n=3), 14% responded to Hep A vaccine alone (n=2), and 64% of patients did not respond to either vaccines (n=9). In the group receiving both vaccines, we analyzed the B cell compartment by flow cytometry. There were no significant differences in the % of core B cell subsets: switched memory, double negative memory, un-switched memory, and naïve B cells at baseline and 12 and 24 WK after anti-TNF treatment between those responding to at least 1 vaccine (R) and those not responding (NR).  The mean % (SE) of germinal center-like B cells in the R group was significantly higher than the NR group over all visits (GC like: CD19+IgD-CD27-CD38++CD24-) (R: baseline 2.3±0.66, WK12 1.9±0.50, WK24 2.2±0.44 vs. NR: baseline 0.9±0.49, WK12 0.8±0.39, WK24 0.6±0.33; p=0.04 overall mean, 0.01 WK24).

Conclusion: These data suggest that the in vivo generation of B cell memory in response to the T dependent neoantigens hepatitis B and hepatitis A is impaired in RA patients treated with anti-TNF. The % GC B cells in the periphery may correlate with responses to vaccination.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/decreased-vaccine-responses-in-rheumatoid-arthritis-patients-receiving-anti-tnf-treatment-and-relationship-to-b-cell-subsets/